Abstract The overall goal of this proposal is to quantify independent effects of sunlight exposure and psychological stress on recurrence of clinical ocular and nonocular herpes simplex virus (HSV) infections. The aims of this proposal are to: (1) estimate the total (i.e. direct and indirect) effect of unprotected sunlight exposure on risk of HSV recurrence; and (2) estimate the total effect of psychological stress on risk of HSV recurrence. We hypothesize that: (1) unprotected sunlight exposure will exhibit a positive dose-response association with HSV recurrence and that (2) acute psychological stress will be associated with increased risk HSV recurrence. To accomplish these aims, data from 308 adults enrolled in a substudy formed to explore the triggers for HSV recurrence will be reanalyzed. This substudy recruited participants between 1992 and 1996 from a randomized control trial administered by the Herpetic Eye Disease Study group and funded by the US National Eye Institute to test prophylactic antiviral therapy for prevention of HSV recurrence. The 308 participants were seen at regularly scheduled visits at one of 74 clinical sites in the US; additionally, participants provided weekly diaries reporting on sunlight exposure, psychological stress, and other variables. Prior analyses of data collected in this substudy did not adjust for time varying confounding, and suffered from poor assessment of sunlight exposure. Moreover, none of the existing published reports on triggers of HSV recurrence have accounted for the effects of time varying confounding. Standard methods for statistical analysis (e.g. restriction, stratification, regression) may fail to estimate the total effect of time varying exposures/triggers on HSV recurrence in the presence of confounders that vary in time. To ameliorate this limitation, we will use marginal structural models which disentangle the effects of multiple time varying triggers of HSV recurrence. Also, a sunlight exposure metric that combines self-reported exposure and ultraviolet (UV) radiation index as obtained from the National Oceanic and Atmospheric Administration will be created to more finely delineate differences in sunlight exposure. Weekly averaged UV index values will be obtained for each of the 74 geo-coded clinical sites. Results from this proposal will further the understanding of the determinants of HSV recurrence and provide additional evidence for (1) effective behavioral counseling for individuals infected with HSV and (2) development of environmental controls (e.g., sunglasses designed to block critical UV wavelengths or UV blocking eye drops) to reduce morbidity. (e.g., sunglasses designed to block critical UV wavelengths or UV blocking eye drops)